Núria Armengol Urpí

nuria_photo.jpeg


Hi! I am Núria. I recently completed my PhD at ETH Zurich in the Computational Robotics Lab, co-supervised by Prof. Stelian Coros at ETH and Prof. Georg Martius from the Autonomous Learning Group at the University of Tübingen. My PhD was partially funded by the CLS doctoral fellowship. Before that, I earned my MSc in Robotics, Systems and Control from ETH Zurich, completing my Master’s thesis in the LAS group supervised by Prof. Andreas Krause.

My research focuses on sample-efficient sequential decision-making for robotics. I aim to develop algorithms that leverage diverse inductive biases to empower agents to efficiently explore new environments and autonomously navigate the real world. My recent work centers on zero-shot RL, specifically on active exploration for learning behavior foundation models from online data in the most sample-efficient way.

In my free time, I love playing sports, reading, and playing music.

news

May 15, 2026 :robot: FB-MEBE (Maximum Entropy Behavior Exploration for Sim2Real Zero-Shot Reinforcement Learning) accepted in RLC 2026!
May 01, 2026 🗂️ RL-BREW I am co-organising the 3th Workshop in Reinforcement Learning Beyond REWards in RL Conference in Montreal (15th August 2026).
Nov 28, 2025 👩🏽‍💻 Finshed internship at Amazon AWS in Berlin, working on agentic LLM’s.
Sep 17, 2025 🗂️ EWRL 18 I am co-organising the 18th European Workshop in Reinforcement Learning (EWRL18) in Tübingen.
May 28, 2025 🐝 FBEE (Epistemically-guided forward-backward exploration) accepted in RLC 2025!

selected publications

  1. FB-Teaser.png
    Maximum Entropy Behavior Exploration for Sim2Real Zero-Shot Reinforcement Learning
    Jiajun Hu, Núria Armengol Urpí, Jin Cheng, and 1 more author
    2026
  2. fbee.png
    Epistemically-guided forward-backward exploration
    Núria Armengol Urpí, Marin Vlastelica, Georg Martius, and 1 more author
    In Reinforcement Learning Conference, 2025
  3. caiac-1.png
    Causal Action Influence Aware Counterfactual Data Augmentation
    Núria Armengol Urpí, Marco Bagatella, Marin Vlastelica, and 1 more author
    In Forty-first International Conference on Machine Learning, 2024
  4. elfp.png
    Efficient Learning of High Level Plans from Play
    Núria Armengol Urpí, Marco Bagatella, Otmar Hilliges, and 2 more authors
    In International Conference on Robotics and Automation, 2023
  5. oraac_2.png
    Risk-Averse Offline Reinforcement Learning
    Núria Armengol Urpí, Sebastian Curi, and Andreas Krause
    In International Conference on Learning Representations, 2021